Depends on Quast version 3.2
Pandoc (version > 1.17) must be installed
1.Clone this repository with git clone https://github.com/pbelmann/quast-evaluator.git
2.Change working directory: cd quast-evaluator
3.Run R -e 'install.packages("packrat" , repos="http://cran.us.r-project.org"); packrat::restore()'
Quast Evaluator has two modes:
This mode lets you collect quast results from different quast output directories.
Run
Rscript quast_evaluator_cli.r new /path/to/assemblers.tsv /path/to/info.tsv -o /path/to/output/directory
where assemblers.tsv and info.tsv must have the following columns:
- assemblers.tsv:
name assembler path group
assembler1 jolly_euclid_1 /home/belmann/projects/quast-evaluator/tests/testthat/data/assembler1 1
assembler2 angry_newton_2 /home/belmann/projects/quast-evaluator/tests/testthat/data/assembler2 2where
-
/home/belmann/projects/quast-evaluator/tests/testthat/data/assembler1must contain the directoriesruns_per_referenceandcombined_reference -
nameis the assembler name that will be shown in the plots -
assembleris the name of the assembler provided to metaquast. -
info.tsv:
id path group length mapping
1030752.gt1kb 1030752.gt1kb strain 1200 43000
1030755.gt1kb 1030755.gt1kb strain 2300 89999
1030836.gt1kb 1030836.gt1kb evolved 4500 76666
1030878.gt1kb 1030878.gt1kb evolved 8999 50000where
pathis a directory inside theruns_per_referencedirectory.idis a uniq identifiergroupis an identfier which groups different referenceslengthis the length of the referencemappinghow many reads are mapped to the specific reference
This mode lets you reuse tsv files produced by the 'new' mode.
Run
Rscript quast_evaluator_cli.r reuse /path/to/combined_ref_data.tsv /path/to/ref_data.tsv -o /path/to/output/directory
where combined_ref_data.tsv and ref_data.tsv have the columns stated in these example files: tests/testthat/data/ref_data.tsv and tests/testthat/data/combined_ref_data.tsv
- '-c /path/to/plot.conf' This file allows you to define min and max for each plot individually. The here for an example.
See tests
You can run tests with
R -e 'devtools::test()'Please use this if you want to propose a Pull Request.